Osteoporosis is the most prevalent metabolic bone disease and a major public health problem mainly characterized by low bone mineral density (BMD). BMD has a heritability > 60%.The specific genes involved are argely unknown. Women have lower BMD and higher risk to osteoporosis than men. Our previous studies have demonstrated that some osteoporosis risk genes/genomic regions are gender specific. The GOAL of this project is primarily to identify such osteoporosis genes for females and, secondarily, to assess the gender specificity of these identified genes in male samples. Potential none-genetic covariates and nteractions will be assessed and significant ones will be adjusted for. Using unrelated Caucasian female samples that we have accumulated in the past 10 years, we propose to conduct a powerful genome wide association (WGA) scan for BMD genes important for females. Our earlier data obtained in whole genome linkage scans (WGLS) and meta-analyses, DMA micoarray and proteomics studies, and those to be obtained in Projects 2 and 3 of this SCOR will be used to guide focused analyses of this WGA. The 300 most significant genes/genomic regions identified in the WGA will be followed for validation in 800 nuclear families (each with two parents and at least two offspring aged 25-45) that we have recruited by the support of R01AR050496. Those genes that remain significant after transmission disequilibrium test (TDT) in the female offspring (n=936) will be tested by TDT in the male offspring (n=714). Those genes that remain significant in the males are common for risk of osteoporosis in both sexes; otherwise, they are female specific. Our hypothesis is: sex-specific genes for BMD variation can be detected with a powerful WGA and robust TDT when complemented by previous WGLS and our gene and/or protein expression studies in major bone cells. We will fulfill the following Specific Aims: 1) To perform a powerful WGA study using latest Affymetrix SNP chips for 400 healthy women with high and 400 osteoporotic women with low BMD (belonging to aged matched population top or bottom 20%, respectively); 2) To compare the results obtained from the WGA with WGLS and gene/protein expression data (including those to be obtained in Projects 2 and 3 of this SCOR) and other available data of in vivo and in vitro studies; 3) To evaluate the 300 most promising genes/genomic regions obtained through Specific Aims 1 and 2 using ~10,000 SNPs in the 800 nuclear families with robust association analyses for female offspring and subsequently in male offspring, to identify sex-common and female-specific BMD genes; 4) To evaluate the most promising markers obtained through Specific Aim 3 in other populations, including US Caucasians, Blacks, one Israel population, one Amish Jewish population in US, Mexican Americans and Han Chinese. Identifying genes for human BMD variation, especially for women, is important for 1) gaining insights into the fundamental molecular mechanisms of risk to osteoporosis, 2) discovering new pathways and targets for therapeutic cures; 3) identifying genetically susceptible individuals, so that future preventions and interventions can be targeted to and based on individuals' specific genotypes.